GIGABYTE Latest Beta BIOS

[Supernaut]

I'm getting pretty frustrated trying to hit 1900 FCLK on this 5900X. Tried several RAM kits today all on stock settings and still can't get a POST. I'm not sure how to determine if it's the CPU, the BIOS or the motherboard at this point without spending a ton of money on gear that is for the most part unavailable right now... has anyone had success on F33a yet at 1866+? Aorus Xtreme, btw. Gigabyte hasn't responded to my ticket yet, it's been about a week.

I can hit 3800mhz memory and 1900fclk easy. It can do 4000 and 2000fclk without even changing cas timings on memory but I get USB stutter at 2000fclk. Currently just running at 3800mhz and 1900fclk.

SOC voltage 1.2
DRAM voltage 1.44

Ram kit -

• TDPPD432G3600HC16GDC01

[Montano5]

I'm getting pretty frustrated trying to hit 1900 FCLK on this 5900X. Tried several RAM kits today all on stock settings and still can't get a POST. I'm not sure how to determine if it's the CPU, the BIOS or the motherboard at this point without spending a ton of money on gear that is for the most part unavailable right now... has anyone had success on F33a yet at 1866+? Aorus Xtreme, btw. Gigabyte hasn't responded to my ticket yet, it's been about a week.

I'm at 1867 Mhz (MCLK / FCLK / UCLK) and F33A works like a charm. I can't have better results with 1900. At 2000, the motherboard POST but with many WHEA. So I didn't look further.

I noticed Gigabyte set the SOC voltage at 1.17 / 1.18v on AUTO. it's a bit too high according to me. Try to set it at 1.05 or 1.1 manually.

stasio : Thank you for sharing all the files with us ! You do a great job ;) Do you think Gigabyte could allow tRRDS minimum value to 3 ? (The lower value I can't set is 4 actually on my Gigabyte AORUS Elite).

[FedeS]

Yes. I use qbittorent. And tested it a lot. Everything is stable for me now. Try different options.

Ok i solve all problems with driver 10.46 and disable this advanced settings according to MSI and Asrock forums:

Advanced EEE
ARP Offload
Energy-Efficient Ethernet
Gigabit-Lite
Green Ethernet
Interrupt Moderation
IPv5 Checksum Offload
Large Send Offload v2 (IPv4)
Large Send Offload v2 (IPv6)
NS Offload
Power Saving Mode
TCP Checksum Offload (IPv4)
TCP Checksum Offload (IPv6)
UDP Checksum Offload (IPv4)
UDP Checksum Offload (IPv6)

Next setp contact Gigabyte support for ITE Device issue (set_address_failure) on B550i Aorus Pro

[AMD718]

Well your Bus Speed (BCLK) is awful low. 98.7 MHz. Never saw one run that low. You're losing an awful lot of performance. That's the one thing you want at 100MHz or higher. Makes a huge difference the faster you can get data running around the mobo. Could be why you're getting these 'Bus/Interconnect Errors'. Your data is just crawling along.

The BCLK is always reported a few MHz below the 100 MHz set in the BIOS. Not sure how you arrive at losing an awful lot of performance. My C20, C23, CPU-Z, GB4, GB5, Aida, etc., benchmarks all seem to be in-line or higher than most 5950x I've observed. From the single Aida64 screenshot I posted, what do you see is "crawling along"? The WHEA errors at IF2000 appear to be pretty common from what I gather. I can decrease their frequency with additional VDDP and VDDG but cannot eliminate them entirely.

[AMD718]

Well your Bus Speed (BCLK) is awful low. 98.7 MHz. Never saw one run that low. You're losing an awful lot of performance. That's the one thing you want at 100MHz or higher. Makes a huge difference the faster you can get data running around the mobo. Could be why you're getting these 'Bus/Interconnect Errors'. Your data is just crawling along.

This is also normal, and the result of spread spectrum "spread spectrum - clock drifts between 100% and 99.5% of the set value. This avoids creating a lot of electromagnetic noise on a single frequency at the cost of 0.25% of performance. Some boards provide an option to disable it, some don't. Modern systems are often less stable with it disabled."

[Coolsys]

All these USB issues may not only be Gigabyte but only Gigabyte BIOS updates can help us

The ITE device is a chip that handles many functions and normally is accessed directly from the CPU (see block diagram in the manual).
To be seen by RGB Fusion software it is likely also connected to USB and this is failing after a restart
Its possible that USB access was added or messed up in F10 as the issue isn't mentioned before

Noctua PWM CPU fans dropping to zero speed during Windows load is a known issue but Gigabyte have not fixed it

The AX200 bluetooth module needing to be unplugged issue is known across many brands for a long time
Also the Realtek Ethernet disconnect issue is reported across many brands so it was a poor choice by Gigabyte to still use it

Hi, I wanted to let you know that I have experienced something similar, maybe worse. But I could only find this issue happening with Noctua NF-A12x25 Fans on CPU_FAN, CPU_OPT and SYS_FAN2 header, though the last mostly works. I investigated further and found to resolve this issue by setting control mode to "voltage" is key. Also the fan stops earlier, in my test during BIOS reinitialization and can't be started until unplugged and re-plugged, or power cycle of mainboard. Most likely a fan stop occurs after a setting in the BIOS was changed and saved (and the restart after this, of course).

The downside of a stopped fan is, that the fan will draw full maximal current. I noticed the stopped fan by smelling "electronics" and wondering where that might come from. Fortunately the fan (or better none of the ones that were stopped, which are 4 of 6 Fans I own and use for my watercooling) did not take damage.

Can you confirm, that you are reporting the issue with the same fan model? I got an answer from Noctua that they are aware of this issue and are investigating, but given the time this fan is on the market, I doubt they will be able to resolve this in collaboration with Gigabyte. Gigabyte got a service request from my side as well.

[jwsg]

Hi, I wanted to let you know that I have experienced something similar, maybe worse. But I could only find this issue happening with Noctua NF-A12x25 Fans on CPU_FAN, CPU_OPT and SYS_FAN2 header, though the last mostly works. I investigated further and found to resolve this issue by setting control mode to "voltage" is key. Also the fan stops earlier, in my test during BIOS reinitialization and can't be started until unplugged and re-plugged, or power cycle of mainboard. Most likely a fan stop occurs after a setting in the BIOS was changed and saved (and the restart after this, of course).

The downside of a stopped fan is, that the fan will draw full maximal current. I noticed the stopped fan by smelling "electronics" and wondering where that might come from. Fortunately the fan (or better none of the ones that were stopped, which are 4 of 6 Fans I own and use for my watercooling) did not take damage.

Can you confirm, that you are reporting the issue with the same fan model? I got an answer from Noctua that they are aware of this issue and are investigating, but given the time this fan is on the market, I doubt they will be able to resolve this in collaboration with Gigabyte. Gigabyte got a service request from my side as well.

I'm running a NF-A14 as CPU fan and noticed it stopping for a few seconds only as Windows loads but since I put the case side back on obviously I dont see it now - and I think many people dont realise this is happening to them too.

Ive seen the suggestion to use voltage control but Ive not seen a CPU fan under voltage control rather than PWM for many years - on my previous Asus m/b it wasnt even an option for the CPU fan. It's bad this hasnt been fixed yet.

But your symptoms of multiple fans all stopping and smelling of electronics sounds more like fan stall than a normal PWM fan stop and maybe there is some other issue with your setup.

[T.kla]

Did anyone measure the real SOC voltage on the mainboard electrically?
I think the voltage could be higher then stated in UEFI.
My B550 Elite Ax V2 set it automatically to 1.2v with 3600mhz ram
i then manually lowered it until 0,95 which resulted in a crash but running stable as low as 0,98v. Also with 3600mhz ram
i think due to stability problems this voltage may be higher than stated.
so, if u run 1,2v maybe in reality it is 1,3v or more?

[Rock3t]

Ok i solve all problems with driver 10.46 and disable this advanced settings according to MSI and Asrock forums:

Advanced EEE
ARP Offload
Energy-Efficient Ethernet
Gigabit-Lite
Green Ethernet
Interrupt Moderation
IPv5 Checksum Offload
Large Send Offload v2 (IPv4)
Large Send Offload v2 (IPv6)
NS Offload
Power Saving Mode
TCP Checksum Offload (IPv4)
TCP Checksum Offload (IPv6)
UDP Checksum Offload (IPv4)
UDP Checksum Offload (IPv6)

Next setp contact Gigabyte support for ITE Device issue (set_address_failure) on B550i Aorus Pro

I think you turned off more than it requied to get it stable, but it is good to hear you solved the problem.

[SrKag]

Hi first time in here. 3700x on a Master v1.0. I don't want to overclock cpu but i am working on RAM tweaks. RAM runs good at CL 15 3800/3800 1 to 1.
Been Looking for pages to help to decipher the Gigabyte BIOS for F33a, Can anyone direct be to a page with this info?
Would love to pick some brains on BIOS tweaking and RAM tweaking. Any takers?

[nicegamer7]

I just wanted to post that the issue of not seeing anything until Windows is booted was caused by my graphics card not supporting DisplayPort 1.4 ????.

In case anyone else is facing the same thing, I fixed it with https://www.nvidia.com/en-us/drivers...fi-update-x64/.

[chinmoy32]

Dear Gigabyte rep, stasio Please read the following and you you could do anything about it I would really appreciate it.

Gigabyte Technical Support Team came through today with the F9k BIOS for Z390 Aorus Xtreme [Fixed Settings for Disabling 10 GIG LAN]

Here is the link https://drive.google.com/file/d/1iHv...yhK9CqUb2/view

[mastah]

Currently doing FCLK 2000, Mem ratio 40, UCLK==MEMCLK with XMP profile for timings (F4-3600C16D-32GTZN).
vSOC at 1.050
vDDG at 1.000

Gigabyte X570 Aorus Xtreme 1.1
F33a

No whea so far.

[jwagriff]

Hi all, just wanted to first of all say thank you to stasio for the continuous updates regarding gigabyte BIOS'. You're a lifesaver man!

I've made an account to ask about an 'issue' im having with my B550M AORUS ELITE on the Official release of the F12 bios. Sometimes, when I turn my PC on, the only thing that displays is a black screen with a white '_' in the top left. The only thing that resolves it is to either reset the PC via the reset switch or turn it off and on again, after which, it boots up perfectly normally.

This '_' usually appears before the gigabyte logo as well as the 'DEL to enter bios' text at the bottom of the screen. I'm wondering if anyone else has encountered this, and if so, how did you fix it? Everything else appears to be working fine with my board, such as USB ports and ethernet, so I see no reason to RMA the thing.

One thing to note is that I think this only started occuring after the F12 update. I was previously on F11a I believe, and I don't think I had any issues on that BIOS (but again, I could be wrong)

The most logical thing for me to do would probably be to flash F13 and see if it still persists, but I kind of want to wait until it's at a full release before doing so. I have tried going through my power settings but, nothing seems to be on or disabled that would make a difference. It's worth pointing out that every time I turn my PC off before the following day, I also turn it off by the mains... could that be a potential cause?

My power supply is an EVGA 650w G2 (Bought in September 2016... maybe that's a potential cause?), I have an R5 3600 and GTX 1060 as well as 16 gigs of Corsair Vengeance Pro RGB, if that helps at all!

Any feedback would be greatly appreciated, thanks all!

[Esormit88]

Lately, my first boot up of the day, my 5900X's first 6 clocks are STUCK at 560MHz and refuse to go higher. Performance is drastically decreased so it's not just a bug. I can't say if I remember this problem BEFORE upgrading to F33a on my X570 Pro Wi-Fi, but it definitely was noticed after the F33a upgrade. When I reboot, it solves the problem and all cores are at max boost. Is this an F33a problem?